Analysis Group - Carbon Control and Competitive Wholesale … · 2018-10-05 · Carbon Control and...

Carbon Control and Competitive Wholesale Electricity Markets: Compliance Paths for Efficient Market Outcomes

Analysis Group

Susan F. Tierney Paul J. Hibbard

May 2015

Acknowledgments

This report presents a review of the potential impact of various intra- and inter-state

Clean Power Plan compliance options on the efficiency of and pricing in wholesale

electricity markets. It also presents implications of the interaction between compliance

strategies and competitive markets for the cost of carbon control across the states. This

is an independent report by Analysis Group, supported by funding from the Electric

Power Supply Association (EPSA). The authors wish to thank EPSA for its support of

the analysis presented in this report. The views and opinions expressed in this study,

however, are solely those of the authors and do not necessarily reflect the views and

opinions of Analysis Group, employees and other affiliates of Analysis Group, EPSA as

an organization, or any particular EPSA member with respect to any issue.

About Analysis Group

Analysis Group provides economic, financial, and business strategy consulting to

leading law firms, corporations, and government agencies. The firm has more than 600

professionals, with offices in Boston, Chicago, Dallas, Denver, Los Angeles, Menlo Park,

New York, San Francisco, Washington, D.C., Montreal, and Beijing.

Analysis Group’s energy and environment practice is distinguished by expertise in

economics, finance, market analysis, regulatory issues, and public policy, as well as

significant experience in environmental economics and energy infrastructure

development. The practice has worked for a wide variety of clients including energy

producers, suppliers and consumers; utilities; regulatory commissions and other public

agencies; tribal governments; power system operators; foundations; financial

institutions; and start-up companies, among others.

Analysis Group

Analysis Group

Table of Contents

EXECUTIVE SUMMARY 1

1. EPA RULES FOR CARBON EMISSIONS FROM POWER PLANTS 4

Introduction: EPA’s Proposals for New and Existing Electric Generating Units 4

The Context for Regulating Emissions from Power Plants 5

CO2 Emission Control for New and Existing Power Plants: States’ Compliance

Choices Matter 7

2. WHOLESALE ELECTRICITY MARKETS 13

Overview: Organized Wholesale Power Markets 13

How Emission-Control Costs Factor into Asset Decisions and Offers in Competitive

Wholesale Markets 15

3. ALIGNING STATE PLANS WITH POWER MARKET STRUCTURES: CREATING INCENTIVES FOR LOWEST-COST CO2 COMPLIANCE AND EFFICIENT POWER MARKET OUTCOMES 22

Why States Should Care About Aligning their CO2 Compliance Plans with

Wholesale Electric Market Structure 22

Benefits of Market-Based Approaches in States’ Clean Power Plans 26

4. CONCLUSION 31

Appendix: Primer on Organized Wholesale Power Markets 34

Carbon Control and Competitive Wholesale Electricity Markets May 2015

Analysis Group Page 1

EXECUTIVE SUMMARY

Acting under its existing authorities under the Clean Air Act (CAA), the U.S. Environmental

Protection Agency (EPA) has been developing proposals designed to reduce carbon dioxide (CO2)

emissions from new and existing fossil-fuel power plants in the United States.

Power production is the nation’s largest source of carbon emissions, contributing

37 percent of all CO2 emissions in the U.S. This will be the first time that federal

policy will broadly address CO2 emissions from the power sector, and EPA’s

new policies will affect over half of the nation’s existing generating capacity.

EPA and the electric sector have a long track record of successful market-based, regional

emission-allowance-trading programs, which can serve as a template for the regulation of CO2

emissions.

Such programs – for example, the Acid Rain sulfur dioxide (SO2) trading

Program, the “NOx SIP Call,” and the Regional Greenhouse Gas Initiative (RGGI)

– provide price signals to power plant owners so that they can make their own

choices about their lowest-cost path to compliance. These programs have proven

to allow for environmental improvements at much lower cost to both power

plant owners and electricity consumers.

There is a natural fit between market-based emission-control programs and competitive wholesale

power markets.

Most of the power plants covered by EPA’s Clean Power Plan operate in

competitive wholesale electric markets administered by an Independent System

Operator (ISO) or Regional Transmission Organization (RTO) (collectively

“RTO”). These regions span more than two-thirds of the states, encompass 70

percent of the nation’s generating capacity, and serve the electricity needs of two-

thirds of the American people.

The CAA allows EPA and the states to establish an emission-allowance trading system for the

control of CO2 from existing and new power plants.

States will ultimately decide how to design their implementation plans

considering state-specific industry structures, circumstances, and policy

objectives. How states choose to implement the rules for existing and new

power plants will affect the competitive conditions as well as overall compliance

costs – positively or negatively – in wholesale power markets around the U.S.



States should take into consideration the potential ramifications of their State Plan designs on

competitive power markets so as to avoid unintended and costly market distortions.

Although the 111(d) rule is focused on existing power plants, the regulation will

alter the competitive landscape for both new and existing power plants. Taking

into consideration both of the upcoming EPA rules under 111(b) and 111(d) of

the CAA, regulating new and existing plants, respectively, the proposed rule

allows states to choose whether to include the new power plants in the program

affecting existing electric generating units (EGUs). We encourage them to do so,

to avoid market distortions. All existing trading programs, including the Acid

Rain trading program and RGGI, include both new and existing sources.

A competitively neutral design of an emission-trading program would be one

where the states in an RTO voluntarily elect to prepare and implement a multi-

state, mass-based plan covering all of the existing generators and new units

within the footprint of the RTO. The overall cap on emissions would be set at the

sum of all of the participating states’ mass-based emissions goals for those EGUs.

With all EGUs in the market facing the same carbon price, system operators

would be able to optimize the dispatch of the generating fleet to minimize both

energy production costs and costs to reduce CO2 emissions.

By contrast, an economically inefficient approach would be one in which at least

some states exclude new generating units from the program. This would lead to

competitive distortions without necessarily producing CO2-emission reductions

or cost savings to consumers. With some states excluding new gas plants from

the program, states would create economic incentives for developers to build

new gas-fired generating units in states where such plants would be excluded

from the mass-based program. The output from these plants would likely to shift

from existing gas-fired units without a net reduction in emissions or costs to

consumers.

There are other, equally bad inefficiencies that could arise with some program

designs, including situations where different states in an RTO adopt mass-based

approaches and others adopt rate-based designs. This would create distortions in

the competitive wholesale market and in the cost per ton of CO2 emissions

reduced.

In short, the design and administration of a new environmental regulation can create competitive

advantages and disadvantages for different power plants and varying financial impacts for

different power plant owners – with cost impacts for consumers.

To design a State Plan that is compatible with the functioning of competitive

power markets (i.e., enabling them to produce and deliver power efficiently and



reliably) and to produce cost-effective CO2-emission reductions, the states in a

common RTO market should support an approach that treats similarly situated

competitors in similar ways, with all of them facing an incentive to control

carbon emissions as economically as possible.

EPA’s Clean Power Plan proposal has given the states broad discretion about how to design their

compliance plans.

Ideally, states will elect to adopt plans that align with the character of the electric

industry structure affecting power plants in different parts of their states. States

should adopt State Plan elements that encourage power plant owners to make

economically-rational decisions. In regions with organized multi-state

wholesale power markets, this would mean that states would proactively work

with their counterparts in the region to develop a common multi-state approach

with similar core elements. Even if they don’t file a single joint multi-state plan,

the states in an RTO should design their plans to be compatible with one another.

There are benefits of a multi-state, mass-based emission-trading program that provides for

emissions-averaging across the footprint of the system as a whole, and folds new units into the

system at the time of their entering commercial operation (or shortly thereafter).

These benefits include: economically efficient power supply; lower-cost CO2

compliance; lower overall costs to electricity consumers; reliability assurance;

administrative simplicity; equity among power suppliers; and environmental

integrity.

We strongly encourage states with electric generating units in RTO power markets to coordinate

the development of their state plans and elect to participate in a common, mass-based multi-state

emission-trading program that covers both existing electric generating units and new fossil

generating facilities.

We hope that states will decide for themselves that this is the core element of a

pathway toward least-cost CO2 compliance and an efficient, reliable power

system for the benefit of their consumers.

We think that a well-designed multi-state mass-based system dovetails with

other state policies and note in particular the findings of so many analyses that

indicate that the overall cost of compliance will be lower with energy efficiency

as a complementary strategy.

We also encourage EPA to provide incentives for states to adopt mass-based approaches that

include both new and existing fossil generating units.



1. EPA RULES FOR CARBON EMISSIONS FROM POWER PLANTS

Introduction: EPA’s Proposals for New and Existing Electric Generating Units

Over the past several years and acting under its existing authorities under Sections

111(b) and 111(d) of the Clean Air Act (CAA), the U.S. Environmental Protection Agency

(EPA) has been developing proposals designed to reduce carbon dioxide (CO2)

emissions from new and existing fossil-fuel power plants in the United States. EPA is

targeting the power sector in part because power production is the nation’s largest

source of carbon emissions, contributing roughly one third of all greenhouse gas (GHG)

emissions in the U.S.1

The current Clean Power Plan proposal was developed under Section 111(d) of the

CAA. Under this section of the CAA, EPA sets emission standards for each state and

provides guidance for state implementation but leaves much of the design and

administration of emission controls to the states. In June 2014, EPA proposed the Clean

Power Plan with state-specific emissions standards for existing fossil-fueled power

plants. The final rule, anticipated in mid-summer 2015, will require each of the 49 states

with affected power plants to prepare and submit a plan for how it proposes to reduce

emissions from the affected generating units in its state.

Although the features of the final regulations will undoubtedly change in light of the

many comments filed, EPA’s current proposal requires states and affected electric

generating units (EGUs) to demonstrate progress toward emission reductions starting in

2020 and with a subsequent reduction after 2030. This new policy will eventually affect

over half of the nation’s existing generating capacity and all but the smallest fossil-fuel

generating units.

The proposed Clean Power Plan came months after EPA published its proposed uniform

national standards under the CAA’s Section 111(b) to limit the amount of CO2 emissions

from certain new generating units that use natural gas or coal to produce electricity.2 To

obtain a New Source Performance Standard (NSPS) permit addressing air emissions in

the future, developers of new natural gas-fired combined-cycle units (NGCCs) and new

coal-fired power plants would need to show that their projects satisfy the new

standards. EPA also expects to finalize these rules during mid-summer 2015.

1 Power generation contributed 37 percent of total CO2 emissions and 31 percent of all GHG emissions in the

U.S. in 2013. See Table ES-1 in EPA, “Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2013,”

April 15, 2015. The next largest sources of CO2 emissions are: transportation (31 percent), industrial (15

percent), residential (6 percent), and commercial (4 percent). 2 With certain exceptions, the proposed 111(b) rule would apply to EGUs commencing construction after the

publication of the proposed regulation in 2014.



The Context for Regulating Emissions from Power Plants

Controlling environmental impacts from power plants is nothing new. For many

decades, the developers and operators of power plants have been responsible for

managing the impact of changing – and often complicated – local, state, and federal

environmental regulatory obligations. Since the CAA was initially enacted in the early

1960s, for example, it has been amended four times.3 Each statutory iteration has altered

the mix of control requirements on criteria pollutants (such as sulfur dioxide (SO2),

nitrogen oxides (NOx), particulate matter) for certain affected electric generating units,

sometimes in major ways. EPA’s and the states’ administration of CAA requirements

has also continuously evolved over time, leading to a wide

variety of air emission-control requirements on affected

units.

The CAA is only one example in a long list of different

federal, state and local laws and regulations addressing the

public health, safety and environmental impacts of electricity generators.4 EPA’s

proposed structure for controlling emissions of CO2 is but the latest iteration of

environmental law and policy that began in earnest several decades ago.

Owners of affected power plants have thus often faced new incremental capital

investments and increased operating costs to comply with evolving health, safety, and

environmental regulations. Power plant owners (and regulators) in turn have needed to

figure out how to recoup incremental investments and operating costs through prices or

rates charged for electricity sales.

Power plant owners recover emission-control costs in very different ways, depending

upon the structure of the electric industry in which they operate. Typically, vertically

integrated investor-owned utilities (IOUs) face cost-of-service regulation with rates set

by state public utility commissions (PUCs); municipal electric utilities, cooperatives, and

federal authorities have rates based on costs pursuant to the decisions of elected or

appointed boards (and in some states, by PUCs).

3 The CAA was first passed in 1963, with amendments in 1967, 1970, 1977, and 1990. Energy Information

Administration (EIA), Electricity Generation and Environmental Externalities: Case Studies, September 1995,

page 17.

4 The list that includes, for example, the National Environmental Policy Act (NEPA), the Clean Water Act,

the Endangered Species Act, the Resource Conservation and Recovery Act, and countless other standards

and requirements with significant variation across states and source categories.

Controlling

environmental impacts

from power plants is

nothing new.



Non-utility owners of merchant generating assets attempt to recover their costs in

bilateral contracts with customers and utilities or in the wholesale markets administered

by RTOs.5 Such merchant plant owners, however, may or may not be able to do so,

depending upon competitive market conditions.

Their ability to do so is greatly affected by the manner in which pollution-control

programs are designed and administered. In the ideal (from an efficient and

competitive wholesale-market point of view), pollution-control requirements could be

administered in a way that achieves comparable cost metrics for similar assets (e.g.,

dollars per unit of pollution reduced or avoided), regardless of technology or fuel used

to generate power.

In practice, this goal is often frustrated by the fact that environmental laws are rarely if

ever enacted with provisions aimed at maximizing both

environmental and competitive-market outcomes. Also,

these environmental laws are administered across states

and regions with quite varied electric-industry and market

structure, technological mix, and asset size and age.6

The successful track record of market-based, regional

emission-allowance trading programs – beginning with the

1990 CAA Amendment’s Title IV cap-and-trade program for SO2 emissions from power

plants – has fundamentally shifted the way that emission-control programs can be

5 See Section 2 for a discussion of RTO competitive wholesale markets, with a primer on market design and

operations in the Appendix.

6 Laws and regulations to control pollution, emissions or discharges often treat different classes of

generating units differently in light of variations in their age, economics, location, readiness of commercially

available control technologies, and so forth. For example, many pollution-control programs apply

prospectively and sometimes grandfather existing EGUs while imposing control requirements on new units.

As pollution-control technologies evolve over time, successive new power plants may face application of

different “best available control” technologies.

Also, different parts of the CAA require that EPA use one type of pollution-control mechanism for one type

of pollutant and another type for a different one. For example, such different approaches might involve

unit-specific technology requirements or emission-rate limitations (e.g., for the Mercury and Air Toxics

Standard (MATS)); consumption or flow limits (or performance standards) (e.g., for NSPS for criteria

pollutants in new power plants); emission-rate averaging or “bubbling” (e.g., across units at a single station,

or among plants owned by a single owner; e.g., for volatile organic compounds and other emissions in

many states’ current State Implementation Plans); state, regional, or national emission caps and allowance-

trading programs (e.g., the national Title IV Acid Rain Program with its cap on SO2 emissions and its

emissions-trading program; the 9-state RGGI program); and pollutant taxes, fees, offsets, and power-plant

operating limits and other restrictions in permits.

Allowance trading

programs have

demonstrated the ability

to overcome many of the

complexities of emission

control program design.



designed and administered. Such an approach aligns well with competitive power

markets and overcomes many of the complexities associated with other emission-control

program designs. Such a program design establishes one value on the margin for a ton

of emissions and similarly affects all generating units covered by the program

(regardless of age, type, location, etc.). In this way, emission-control requirements are

set so as to price emissions on a fair and equal basis across resources that are competing

head to head in energy markets. This creates conditions for cost-effective compliance

without interfering with energy-market dynamics. This approach relies on market

forces rather than administrative decisions to provide signals to generating-unit owners

about their lowest-cost path to compliance and allows for an efficient overall cost of

compliance.7

CO2 Emission Control for New and Existing Power Plants: States’ Compliance

Choices Matter

The framework for control of CO2 emissions from existing and new power plants, as

dictated by the CAA and proposed by EPA, allows for but does not require an emission-

allowance trading system. Instead, states will ultimately decide how to design their

7 For example, the Acid Rain Program “is largely considered a successful cap-and-trade system. By 2007, the

program had achieved its 2010 reduction goal at an estimated cost that was considerably lower than that of

command-and-control regulations, which mandate that each power plant adopt a specific technology to

reduce SO2 emissions or a standard that requires each power plant to emit below a specific fraction of SO2

emissions per unit energy produced.” Juha Siikamäki, Dallas Burtraw, Joseph Maher, and Clayton

Munnings, “The U.S. Environmental Protection Agency’s Acid Rain Program,” November 2012.

http://www.rff.org/RFF/Documents/RFF-Bck-AcidRainProgram.pdf. Also, a recent retrospective review of

various studies of the effectiveness of the SO2-emissions trading policy reviewed actual costs of the program

relative to predicted costs prior to the program’s implementation as well as “how the costs of achieving

environmental objectives through cap and trade compare with those of a “counterfactual” (hypothetical

alternative) command-and-control regulatory approach. “In addition to being less costly than traditional

command-and-control policies would have been, the program’s costs were significantly below estimates

generated by government and industry analysts in the debate leading up to the passage of the CAAA. In

1990, the U.S. Environmental Protection Agency (EPA) estimated the cost of implementing the Acid Rain

Program (with allowance trading) at $6.1 billion. In 1998, the Electric Power Research Institute (EPRI), an

industry organization, and Resources for the Future (RFF), an independent think tank, estimated that total

implementation costs would be $1.7 and $1.1 billion respectively (based in part on actual figures for the first

few years of the program…). In sum, the SO2 allowance-trading system’s actual costs, even if they exceeded

the cost-effective ideal for a cap-and-trade system, were much lower than would have been incurred with a

comparable traditional regulatory approach, and were much lower than the trading system’s predicted

costs. There is broad agreement that the SO2 allowance-trading system provided a compelling

demonstration of the cost advantages of a market-based approach.” Gabriel Chan, Robert Stavins, Robert

Stowe, and Richard Sweeney, “The SO2 Allowance Trading System and the Clean Air Act Amendments of

1990: Reflections on Twenty Years of Policy Innovation,” Harvard Environmental Economics Program,

January 2012. http://www.hks.harvard.edu/fs/rstavins/Monographs_&_Reports/SO2-Brief.pdf.

http://www.rff.org/RFF/Documents/RFF-Bck-AcidRainProgram.pdf

http://www.hks.harvard.edu/fs/rstavins/Monographs_&_Reports/SO2-Brief.pdf



compliance plans considering state-specific industry structures, circumstances, and

policy objectives.

How states choose to implement the rules for existing and new power plants is hugely

important. It will affect the competitive landscape and conditions of wholesale power

markets around the U.S., thus affecting power prices, and it will ultimately affect –

positively or negatively depending on compliance paths chosen – overall CO2

compliance costs. Under some compliance plan approaches, states in a common multi-

state electric market could choose to design their compliance plans in ways that create

distortions in those markets. Other approaches will avoid or minimize such outcomes,

while simultaneously creating conditions for minimizing compliance costs. As we

describe below, these effects are particularly relevant in the parts of the country where

EGUs operate in centrally administered competitive wholesale power markets.

EPA’s proposed Clean Power Plan creates opportunities for states to plan their

compliance strategies to avoid introduction of competitive inequities and inefficiencies.

Indeed, a state’s plan can proactively incorporate design elements that allow for

economically efficient outcomes in both power markets and carbon-control costs.

As states consider their compliance options with that goal in mind, it is important to

take into consideration the implications of implementation choices for both existing and

new power plants, together. By the time a State Plan is implemented, in combination

with Section 111(b)/NSPS standards, it will affect how existing and new power plants

compete (against each other) in electricity markets. In contrast, most discussions of the

impacts of the EPA’s regulations to this point have tended to focus primarily or

exclusively on impacts on existing power-system elements only.

Potential compliance approaches will need to start with the fundamental structure of the

requirements for new and existing sources, which differ as follows:

Existing Units under the Clean Power Plan, CAA Section 111(d): EPA’s proposed Clean

Power Plan establishes state-specific output-based emission-performance standards

(in pounds of CO2 per megawatt-hour (MWh)). Although the proposed standards

are based on EPA’s assessment of each state’s ability to reduce carbon emissions

through an analysis that combines four “building blocks,”8 each state is free to select

8 The four building blocks are: (1) improvements in generating-unit operating efficiency (e.g., heat-rate

improvements); (2) reducing emissions from the most carbon-intensive EGUs through substitution of output

at lower-emitting EGUs (e.g., operating gas-fired power plants at higher capacity factors); (3) reducing

emissions from EGUs by substituting output at power plants with low or no carbon emissions (e.g.,

renewable or nuclear power generation); and (4) reducing emissions from EGUs as a result of use of



how it will comply, provided the state’s Plan ensures that the EGUs in the state meet

the standard. Specifically, EPA has not proposed that each affected EGU meet the

standard; instead, EPA has proposed that each state determine how to meet the

standard on average across all of the EGUs in the state, considering various power

sector CO2–reducing options (including, e.g., demand-side energy efficiency). Each

state may select from a wide range of compliance approaches, may roll new units

into the plan, and may design its approach on its own or in collaboration with other

states.

New Units under CAA Section 111(b):9 To obtain an operating permit under the NSPS

program, certain new fossil-fueled power plants10 will have to meet performance

standards for CO2 emissions. As proposed, new stationary combustion turbines

must be at or below the emissions rate of new NGCC units, at different size

categories. This means that NGCC projects will be able to meet the standards

without the need for add-on emission-control technologies.11 The rule also includes

slightly different limits for fossil fuel-fired utility boilers and integrated gasification

combined cycle (IGCC) units.12

Taking into consideration both of these proposed regulations, states will have the option

to choose whether to fold the new generating units into the program affecting existing

demand-side energy efficiency. While the building-block methodology was developed to calculate state-

specific performance standards in a consistent way, states are not even limited to relying upon the “building

block” measures or approaches as the only ways to achieve compliance.

9 Source: EPA. http://www2.epa.gov/carbon-pollution-standards/fact-sheet-details-about-proposal-new-

sources.

10 EPA’s proposed rule for new fossil fuel-fired EGUs include: utility boilers, integrated gasification

combined cycle (IGCC) units and certain natural gas-fired stationary combustion turbine EGUs that

generate electricity for sale and are larger than 25 megawatts (MW). The rule does not apply to any existing

EGUs or to units undergoing modifications or to reconstructed units. The rule also does not apply to liquid

oil-fired stationary combustion-turbine EGUs; new EGUs that do not burn fossil fuels (e.g., those that burn

biomass only); or low capacity-factor EGUs that sell less than one third of their power to the grid.

http://www2.epa.gov/carbon-pollution-standards/fact-sheet-details-about-proposal-new-sources.

11 Specifically, EPA has proposed standards of 1,000 pounds of CO2 per MWh for larger units. Larger units

are those with fuel input capacity greater than 850 million British Thermal Units per hour (mmBtu/hr)). For

smaller units (those with fuel input capacity less than 850 mmBtu/hr), EPA has proposed a standard of 1,100

lbs CO2/MWh.

12 The proposed limits for fossil fuel-fired utility boilers and IGCC units are based on the performance of a

new efficient coal unit implementing partial carbon capture and storage (CCS). EPA is proposing two limits

for fossil fuel-fired utility boilers and IGCC units, depending on the compliance period that best suits the

unit. These limits require capture of only a portion of the CO2 from the new unit. These proposed limits are:

1,100 lbs CO2/MWh-gross over a 12-operating month period; or 1,000-1,050 lbs CO2/MWh-gross over an 84-

operating month (7-year) period. http://www2.epa.gov/carbon-pollution-standards/fact-sheet-details-about-

proposal-new-sources.

http://www2.epa.gov/carbon-pollution-standards/fact-sheet-details-about-proposal-new-sources





sources once the new units go into operation or at some point after that. This is one of

four types of flexibility that EPA’s Clean Power Plan provides to the states as they

design their plans. Because all of these forms have implications for whether a state plan

aligns well – or poorly – with competitive markets, we describe them here and

summarize them in Table 1.

Table 1

Sources of Flexibility in States’ Design of their State Plans to Comply with

EPA’s Proposed Clean Power Plan

Source of Flexibility: Explanation:

Form of the CO2-

emissions standard

A state may use a rate-based approach (lbs of CO2/MWh from affected EGUs)

or a mass-based approach (a total cap on CO2 emissions from affected EGUs).

Spatial flexibility

A state may develop a common or different set of approaches to reducing CO2

emissions at EGUs in different parts of the state, and they may enter into

multi-state plans with neighboring and non-contiguous states.

Options for the content

of Plan elements

A state may choose from a wide set of options to incorporate into their state’s

Plan.

Scope of units covered

by the State Plan

A state may choose to fold new generating units into the system applied to

existing units at some point after the new ones go into operation, or the state’s

plan may treat old and new units separately.

Form of the CO2-emissions standard: EPA has proposed to allow states to translate the

rate-based performance standard into a mass-based equivalent. That is, EPA has proposed

default methods by which a state can translate the state’s rate-based performance

standard (in lbs of CO2/MWh) into a mass-based total emission limit (in tons per year of

CO2 emitted from the portfolio of affected sources). States will also have the ability to

propose an alternative methodology for converting the rate-based standard into an

emissions cap affecting certain power plants.

A recent report13 examining these alternatives has further categorized the ‘straw man’

options as follows: three alternatives that use a mass-based approach and three others

that use a rate-based approach. (See Table 2.) These different straw proposals also

incorporate a second dimension of the flexibility that EPA has built into the Clean Power

Plan: spatial flexibility. States can design compliance plans that average emission

reductions across separate units within the fence of a generation facility or across all

generating units owned by a single company. Compliance obligations could also be

13 Franz Litz and Jennifer Macedonia, Choosing a Policy Pathway for State 111(d) Plans to Meet State

Objectives, Great Plains Institute and Bipartisan Policy Center, April 24, 2015.



bundled across companies within a state, or states can adopt multi-state approaches, for

example, where integrated companies have operating subsidiaries that cross state lines,

or where neighboring states operate within a single bulk power region or wholesale

electricity market. Finally, a state with generating units in more than one RTO may

develop different compliance approaches for those separate parts of the state and enter

into two sets of multi-state plans.

A third source of flexibility relates to the options available to states as they shape their

Plans. Elements of a State Plan could include reductions in tons emitted within the state

by affected EGUs through such things as: requiring heat-rate improvements;

incorporating permit limits on the operations of certain EGUs; retirement of particular

EGUs; making adjustments in the transmission grid; reductions in demand for electricity

in the state by operating the system at acceptably lower voltage levels; capacity

upgrades at nuclear or hydroelectric plants; installing measures to use less water and

thereby reducing electric-pumping requirements of water supply systems; and so forth.

Scope of units covered by the State Plan: Finally, and as noted previously, states may

choose to develop state plans that address only existing EGUs, limiting new source

requirements to the NSPS under Section 111(b). However, states may instead elect to

wrap into the Clean Power Plan compliance scheme new resources as they come on line

(or, alternatively, after some short period of initial operation). In such a case, new

sources would be subject to the NSPS as part of their initial permitting requirements but

then would also participate in a statewide program to address all emissions from EGUs

– for example, in a mass-based cap-and-trade program covering all sources, with or

without the cap expanded to include the emissions from the new units.



Table 2

Straw-Man Options for States to Design Mass-Based Versus Rate-Based Approaches to Limiting CO2

Emissions from Electric Generating Units (Litz/Macedonia)

Form of CO2

Emissions

Standard

Straw Man

Option Explanation

Mass-Based

Straw Man

Approaches

“Utility Budget

Approach”

“The state allocates shares of its mass-based state emissions budget to each utility

(and other plant-owning entities). This allows the utility to then manage its budget

of allowed tons across its entire fleet of affected electric generating units.”

The approach could be implemented in a single state or across multiple states.

“Plant-Level

Emissions

Budget Trading

Approach”

“A state starts with its allowed emissions budget, or the total number of tons that

may be emitted from all of the affected electric generating units in the state for each

year. Emissions allowances are issued by the state with each allowance

representing an authorization to emit one ton of carbon dioxide. Affected electric

generating units must track and report their covered emissions. At specified

intervals, affected units must turn in sufficient allowances to cover their

emissions.”


“Utility Budget

Approach with

Optional

Trading”

This would borrow from the other two mass-based approaches and would “allow

utilities (and other plant-owning entities) the flexibility to meet an emissions

budget across their fleets, and the option to participate in emissions trading with

others that also choose to participate in trading. The approach could be

implemented in a single state or across multiple states. States could create the

infrastructure for trading to occur across state lines.”

Rate-based

Straw Man

Approaches

“Utility Rate

Approach”

“The state prescribes an emissions rate for each utility (or other plant-owning

entity). Each utility then manages its fleet of covered power plants to meet the

prescribed rate through actions at the plants themselves or activities like energy

efficiency that avoid emissions at the plants through energy savings. The flexibility

inherent in this approach is often referred to as ‘bubbling’ because it creates a

figurative bubble over all of the utility’s affected units, allowing the utility in the

rate-based context to average the emissions performance across all of its affected

units.”

“Rate-Based

Trading

Approach”

This one “entails applying a prescribed emissions rate to all affected units. An

affected unit that generates electricity at an emissions rate that is lower than the

prescribed rate will generate emissions credits. A unit that generates electricity at a

rate that is higher than the prescribed rate will need to use credits to adjust its rate

downward. In this way, units that exceed the prescribed emissions rate can

continue to generate electricity as long as the generation is offset with credits. In

effect, the emission rates are averaged across all units in the system and all unit

owners.”

“Utility Rate

Approach with

Optional

Trading”

This one borrows from the two other rate-based approaches and can “provide

flexibility for owners and operators of affected electric generating units to achieve

the prescribed emissions rate in a least-cost manner. A utility rate approach with

optional trading would allow utilities (and other owners of power plants) the

flexibility to meet an emissions rate across their fleets, and the option to participate

in emissions credit trading with others that also choose to participate in trading.


States—ideally with EPA assistance—could create the infrastructure for trading to

occur across state lines.”

Source: Franz Litz and Jennifer Macedonia, “Choosing a Policy Pathway for State 111(d) Plans to Meet State

Objectives, Great Plains Institute and Bipartisan Policy Center,” April 24, 2015.

These multiple degrees of freedom allow for a great deal of discretion by states in

designing their State Plans. And the many degrees of freedom within and across states



create the potential to introduce new environmental regulations in a manner that either

aligns EGU owners’ and other market participants’ incentives with the economic-

efficiency goals of competitive electricity markets, or creates tensions between the two.

As states consider their options, there are strong economic-efficiency arguments for

taking into consideration the potential ramifications of different state-plan elements on

creating unintended distortions in the competition among different power resources in

the wholesale power markets that supply electricity to the state’s consumers.

2. WHOLESALE ELECTRICITY MARKETS

Overview: Organized Wholesale Power Markets

The EPA’s proposed carbon regulations will affect power plants that operate in many

different contexts around the country. One key attribute of the electric industry that has

developed over the past two decades is the existence of

competitive wholesale markets.14

Today, a significant portion of retail customers’ power is

sourced through wholesale supply. In some parts of the

country (e.g., much of the South and the West), wholesale

supply occurs through bilateral purchases and sales, and

most of the power provided to retail customers comes from

the local utility’s own fleet of power plants. In other parts

of the country, the primary framework for wholesale

supply is a centralized wholesale market (also called an ‘organized’ wholesale market).

14 Historically (and even today), when a regulated utility owned a power plant whose output was for the

benefit of that utility’s own customers, the costs associated with that power plant were recovered directly in

retail customers’ rates regulated by state or local authorities. In such a case, there is no wholesale purchase

or sale of power (i.e., no sale for resale), and the local regulator established the terms and conditions of cost

recovery of capital costs and operating expenses in rates charged to end-use customers. When that power

plant provided power for the benefit of another utility and its customers, that power sale occurred as a

bilateral wholesale transaction, with prices subject to regulation of the Federal Energy Regulatory

Commission (FERC). Traditionally, such transactions were priced at cost. That situation changed starting in

the 1990s, when FERC began to evolve its policies to enable the electric industry to rely on competition

(rather than cost-based regulation) as the means to set prices in transactions where the seller lacked market

power. In the electric industry, “market power is the ability of an electricity supplier to raise prices

profitably above competitive levels and maintain those prices for a significant time. Electricity suppliers

exercising market power force consumers to pay higher electricity prices than they would pay in a

competitive market.” EIA, “The Changing Structure of the Electric Power Industry 2000: An Update,” page

78.

How states design their

Clean Power Plans will

either support or

undermine least-cost

compliance and the

efficiency of competitive

wholesale power

markets.



In an organized wholesale market, an RTO administers a bid-based central energy

market with coordinated dispatch of all generating units in the footprint of the system.

In such organized markets, wholesale electricity is bought and sold at market prices

based on that dispatch, under rules approved by the Federal Energy Regulatory

Commission (FERC).15 (See the Appendix for an explanation of the energy, ancillary

service and capacity markets administered by RTOs in the U.S.)

These RTOs cover power plants in over two-thirds of the states, encompass 70 percent of

the nation’s generating capacity,16 and meet the electricity requirements of two-thirds of

the American people. As shown in Figure 1, the U.S. RTOs are the California

Independent System Operator (CAISO), Electric Reliability Council of Texas (ERCOT),

ISO New England (ISO-NE), Midcontinent Independent System Operator (MISO), New

York Independent System Operator (NYISO), PJM Interconnection (PJM), and

Southwest Power Pool (SPP).

15 RTOs are also responsible for the reliability of the transmission grid and for balancing generation and load

at all times over the RTO’s geographic footprint. RTOs conduct short-and long-term planning to ensure

resource adequacy and transmission-system security, assure the provision of non-discriminatory access to

the grid for owners of power plants, and provide for the design and administration of competitive

wholesale electricity markets. All RTOs carry out their work with the input and advice of market

participants and other stakeholders (including representatives of relevant state regulatory agencies).

16 SNL Financial.

Figure 1:

Organized Wholesale Electricity Market Regions (Source: FERC)



How Emission-Control Costs Factor into Asset Decisions and Offers in

Competitive Wholesale Markets

Overview: Owners of non-utility power plants in wholesale markets recover emission-

control costs in very different ways than traditionally regulated utility companies that

own power plants in those markets (or in the parts of the country outside of an RTO

footprint). Owners of merchant generating assets attempt to recover their costs in

bilateral contracts with customers and/or utilities, or from the organized markets for

capacity, energy and ancillary services. They may not be able to do so, however,

depending upon competitive market conditions and other factors.

Notably, states’ decisions about the design of their State Plans will alter the economics of

competitors in wholesale markets. In an RTO market, the clearing price of power

($/MWh) reflects the offer price of the marginal generator dispatched to meet load in a

particular hour (with the offer price typically reflecting that unit’s variable operating

cost, including fuel and emissions allowances). If the marginal generator is an existing

EGU subject to the Clean Power Plan and had to purchase an emission allowance (under

a mass-based approach) or an emissions credit17 (in a rate-based emission-trading

scheme) to cover its CO2 emissions, the clearing price of power will go up, reflecting the

carbon-intensity of that EGU. That clearing price would be paid to generators that are

producing power in that hour, and a generator not dispatched (or not operating) in that

hour would receive no energy payment.

From the perspective of electric energy markets, the design of CO2-emission-control

programs can have a number of different types of impacts on different plants,

depending upon where they are in the supply stack and how the State’s approach

creates incentives for responses by an owner of a power plant. Some of those potential

approaches – e.g., a market-based trading program – price compliance equally across

competitors based on the direct carbon intensity of different EGUs and can be

competitively neutral in energy markets (with the only change being the carbon-

intensity of marginal generation). But other approaches may not. To highlight the

differences among approaches, we describe market impacts of three potential

compliance pathways.

17 Depending on the level of the lbs/MWh performance standard and the emission rate of the marginal

generator, the operating costs of the unit may increase or decrease. If the unit emits above the level of the

standard, its costs per MWh would increase. If the unit emits below the level of the standard, its costs per

MWh would decrease because it earns a credit for each MWh of electricity it produces.



First example: an economically efficient plan for CO2 compliance by EGUs in RTO

states: Consider a situation in which some (or preferably all) of the states in an RTO

elect to prepare and implement a multi-state, mass-based plan covering all of the EGUs –

existing and new – in the footprint of the RTO into a single CO2 program. The overall

cap on emissions would be set at the sum of all of the participating states’ mass-based

emissions goals for those EGUs in that RTO.18 All of the EGUs in the entire footprint of

the RTO would face the same carbon price (in $/ton emitted), in the form of having to

own or control an allowance to emit a ton of carbon.

In such a market, a relatively carbon-intensive EGU would face increased operating

costs (in $/MWh), which might cause the plant to shift to a different (higher-cost)

position on the dispatch stack. That impact might or might not cause it to be dispatched

less often. For example, if the plant were a relatively efficient coal plant that remained

an economical base-load EGU even after the purchase of emissions allowances, and the

increase in operating costs did not cause it to be more expensive than the marginal

generator in any hour, then it would not operate less often. Its operating costs would go

up, but so would its revenues (because of the impact of having higher clearing prices in

the market). By contrast, a less-efficient coal plant or gas-fired power plant which

similarly had to purchase allowances to cover its emissions might end up operating less

because its relative carbon intensity and inefficiency caused it to be more expensive in

some hours than the marginal generator would be in the absence of the CO2-control

program. Overall, the system’s re-dispatch would reflect efficient operations taking

carbon costs into account. The overall electric system would be producing electricity

with lower CO2 emissions and an economic dispatch of competing sources of power

with different carbon intensity.

Second example: an economically inefficient plan for CO2 compliance by EGUs in

RTO states: By contrast, let’s assume that the same states adopt that same plan for

existing units, but some of those states (the “A Group” of states) decide that they want

to keep new NGCCs out of their states’ program and other states (the “B Group” of

states) decide to fold new NGCCs into the program that covers existing units. Finally,

for the states (the A Group) that do not fold new NGCCs into their emissions-trading

18 In instances where a state has EGUs in different RTOs or has some power plants in an RTO and other

EGUs not in an RTO at all, and where the state wants to adopt a state plan that differentiates compliance

elements consistent with those electric-system constructs, then the state will need to establish the emissions

targets for those EGUs in the RTOs. The state could do this, for example, by assigning the state’s average

rate-based emissions target to each EGU and then converting those amounts to a mass-based number for the

EGUs in the RTO.



program, those new plants otherwise have no CO2-emissions requirements beyond what

they needed to obtain their original air permits under the NSPS program.

Would this combination of state plans lead to competitively neutral impacts and

incentives in this RTO market, with the only adjustment reflecting the carbon intensity

of EGUs? No. Here is why: Identical NGCC facilities in the same RTO would face

different price signals, leading to market distortions and a suboptimal allocation of

capital. Existing EGUs in the RTO (across the A and B Groups of states) would have to

buy CO2-emission allowances and include their value in offer prices into the RTO’s

energy markets. A new NGCC located in an A Group state and having a very similar

CO2-emissions profile (i.e., lbs of CO2/MWh) as some of those existing EGUs in the RTO,

however, would not have to. This means the new NGCC with essentially the same heat

rate and operating cost as other existing EGUs would routinely dispatch ahead of others,

increasing its output, its revenues, and its profits relative to other similarly situated

EGUs. The new NGCC would also get the revenue benefit from higher clearing prices

without incurring additional costs. This would give the new NGCC a competitive

advantage over the other generators in the RTO due to state policy design and not

because of any intrinsic cost advantage of the new NGCC.

Moreover, this advantage may actually induce new NGCCs to enter the market in one

of the A Group of states. A new NGCC would have no such advantage if it were located

in a state in the B Group. In fact, a proposal to site and permit a new NGCC in a B

Group state would increase competition for the supply of CO2 allowances for all EGUs

in the RTO, increasing carbon prices and further aggravating the competitive inequities

and market distortions created by the disparate policy approaches. The B Group states

would have the strongest overall limit of CO2 emissions from power production, shifting

emissions to the states in the A Group.

An analysis performed by Calpine Corporation suggests that at certain levels of CO2-

emission-allowance prices, the advantage afforded to new units by A Group states (i.e.,

states that elect to allow a new NGCC to operate outside of the cap-and-trade program

established for EGUs in the same state) could lead developers to anticipate high-enough

revenues in future electric-energy markets to make it worthwhile to invest in the capital

costs of a new power plant.19 This state policy design could lead to more development

19 Calpine’s estimate is based on conditions in the PJM RTO market, where the all-in, levelized cost of new

entry for an NGCC is $31/MWh. In this analysis, Calpine calculated that in the absence of a mass-based



of NGCCs and potential displacement of otherwise economical and low-emitting

generating assets without a commensurate reduction in CO2 emissions. It would lead to

stranded costs at existing units, again, without a CO2 benefit. Such an outcome would

be antithetical to the principles of competitive electric markets and to sound CO2-

emissions-reduction policies.

Third example: another economically inefficient plan for CO2 compliance by EGUs

in RTO states: In this example, let us assume that the states with EGUs participating in

the RTO market all decide to adopt a rate-based CO2-emissions-reduction approach for

those EGUs. This is similar to the “Rate-Based Trading” straw-man approach20

described in Table 2, above.

The EPA’s proposed Clean Power Plan has very different emission-rate targets across

the states. In some cases, neighboring states within an RTO region have significantly

different rate-based targets. Figure 2 shows the state-specific interim (2020-2029) rate-

based targets for each of the 49 states with EGUs. States are ranked from highest to

lowest in terms of the amount of CO2 their targets would allow EGUs in their states to

emit. The highlighted states have EGUs that are members of the PJM RTO. The range in

targets among these states is between 1,844 lbs/MWh (for EGUs in Kentucky) to 647

lbs/MWh (for EGUs in New Jersey).

compliance plan covering all fossil units, an existing NGCC and a new one with equivalent heat rates and

emissions rates would each have the potential to generate revenues in PJM’s energy markets of $20/MWh

above variable operating costs, so that these revenues would contribute to recovery of fixed costs of the

power plants. With a mass-based compliance plan with CO2-allowance prices equivalent to $30/ton, the

impact on clearing prices and energy-market revenues for a new NGCC that did not need to acquire

emissions allowances would equate to approximately $11/MWh. At this level ($20/MWh plus $11/MWh),

the new NGCC would find it economically attractive to enter the market and compete with existing

generators with exactly the same emissions profile. Thus, as a result of state policy, there would be new

entry without any commensurate reduction in CO2 emissions. Calpine Corporation, “The Impacts of the

Clean Power Plan on Wholesale Power Markets.” February 24, 2015.

20 This approach “…entails applying a prescribed emissions rate to all affected units. An affected unit that

generates electricity at an emissions rate that is lower than the prescribed rate will generate emissions

credits. A unit that generates electricity at a rate that is higher than the prescribed rate will need to use

credits to adjust its rate downward. In this way, units that exceed the prescribed emissions rate can continue

to generate electricity as long as the generation is offset with credits. In effect, the emission rates are

averaged across all units in the system and all unit owners.” Source: Franz Litz and Jennifer Macedonia,

“Choosing a Policy Pathway for State 111(d) Plans to Meet State Objectives, Great Plains Institute and

Bipartisan Policy Center,” April 24, 2015.



Figure 2

Interim Period (2020-2029) Rate-Based Targets by State

Source: EPA. http://www2.epa.gov/carbon-pollution-standards/clean-power-plan-proposed-rule-technical-

documents-spreadsheets

Let us assume that Pennsylvania, New Jersey, Ohio, and Virginia all decide to adopt a

rate-based trading approach. Under EPA’s proposal, each state could elect to keep its

own CO2/MWh rate or create a blended average CO2/MWh rate across the other states,

adopting a common multi-state, rate-based approach. Let us assume further that

Pennsylvania and Ohio fold new NGCCs into their systems for existing units, while

Virginia and New Jersey do not.

In this scenario and before the Clean Power Plan, the NGCCs would have similar offer

prices and each would have a similar opportunity to be dispatched in the RTO’s energy



market. But under a Clean Power Plan scenario in which the four states retain their

separate rate-based targets, there would be distortions in the PJM power market. The

NGCCs located in Ohio and Pennsylvania would have the ability to generate an

emissions-reduction credit for every MWh they generate because those NGCCs’

emissions (assumed to be 1,000 lbs of CO2/MWh) are below each of those state’s target

emissions (Ohio at 1,452 lbs/MWh and Pennsylvania at 1,179 lbs/MWh).

Thanks to those generated emissions-reduction credits, the Ohio and Pennsylvania

NGCCs would be able to offer a lower price into PJM’s market than the NGCCs located

in New Jersey and in Virginia, with the latter actually having to purchase emissions-

reduction credits because their emissions (at 1,000 lbs/MWh) are higher than those

states’ target rates (Virginia at 884 lbs/MWh and New Jersey at 647 lbs/MWh). The

NGCCs in Ohio would be able to create the most credits (because their emissions targets

are highest) and their compliance costs would be cheaper than an identical unit in

Pennsylvania, which would further distort the dispatch stack. A New Jersey NGCC

would be the most disadvantaged, and a NGCC in Ohio would be most advantaged. A

new NGCC would be particularly attracted to locating in Ohio or Pennsylvania and

could displace output in the other states without any commensurate reduction in total CO2

emissions. Further distortions would occur if additional PJM states also adopted rate-

based trading programs or mass-based trading programs (e.g., Maryland and Delaware

are part of the RGGI program and part of PJM).

Insights from the examples: State CO2 plans can – but do not need to – create

distortions in wholesale power markets

As these examples show, the design of emission-control programs can affect similar

generating facilities in distinctly different ways, either increasing or decreasing their

operating costs within the same wholesale power market. Consequently, depending on

the form of emission-compliance requirements, these requirements can affect clearing

prices (upward or downward) in energy markets. The character of emission-control

requirements can also affect a particular generating unit’s net energy-market revenues

because that unit may be dispatched more or less often than without those requirements,

its own costs may change, and the clearing prices it receives may be different.

These various impacts mean that the design and administration of a new environmental

regulation can create competitive advantages and disadvantages for different power

plants (and varying financial impacts on the owners of different plants) without creating

a clear or reasonable nexus between the differences in wholesale market impacts and the



purpose of the regulation.21 This, of course, tends to impact power plant owners’

attitudes about whether the new regulation is a good idea or not and about what the

appropriate design of the regulation should be. But it can also affect the ultimate

economic efficiency of wholesale market operations (for example, by improving or

harming heat rates or by facilitating or impeding dispatch of power plants across the

states in a common RTO market). These inefficiencies would lead to adverse impacts on

consumers.

To the extent that a state seeks to design a State Plan with neutral effect on the

functioning of competitive power markets – i.e., enabling them to produce and deliver

power efficiently and reliably and produce cost-effective CO2-emissions reductions – the

State Plan should focus on an approach that treats similarly situated competitors in a

common regional power market in similar ways, with all of them facing an incentive to

control carbon emissions as economically as possible.

If states in a single RTO do not coordinate the development of their State Plans, and

their individual plans differ in significant ways, power plants in that market with the

same – or very similar – technology, fuel, power-production efficiency and emissions

rate will likely end up with quite different dispatch profiles only by virtue of the fact

that they are located in different states, and without any lower CO2 emissions.

The complex interactions between compliance requirements and how such requirements

flow through wholesale markets – with consequences for unit operations, unit additions

and retirements, unit costs, and consequently offers in competitive wholesale energy,

ancillary service and capacity markets – have important implications for wholesale

market outcomes. While the complexity of these various potential combinations of

circumstances within and across states may make it seem as if a workable solution is

untenable, just the opposite is true. The answer or solution is fairly straightforward:

program designs that increase the uniformity of compliance-cost price signals – across

existing and new units, across companies, and across states – tend to eliminate

distortions and inequities in competitive energy markets, support competitive and

efficient energy market outcomes, and lower total compliance costs across the board.

21 This fact is not unique to environmental regulations: This also occurs when other public policies

differentially affect plant investment and operations: a plant with a large workforce will face large financial

impacts of a law changing labor compensation or worker safety issues; a plant with a larger land area will

face a bigger impact if there is a change in property tax; and so forth. Also, changes in fundamental market

conditions (e.g., changes in the relative prices of fuels like natural gas and coal) can – and do – introduce

these differential impacts on power plants as well.



That outcome would be good for electricity consumers.

It will help if EPA and the states develop strategies that support alignment of carbon-

emission controls with competitive market outcomes.

3. ALIGNING STATE PLANS WITH POWER MARKET STRUCTURES:

CREATING INCENTIVES FOR LOWEST-COST CO2 COMPLIANCE AND

EFFICIENT POWER MARKET OUTCOMES

To support economically efficient compliance approaches along with economically

efficient wholesale power market outcomes, states should adopt State Plan elements that

encourage EGU owners to make economically rational decisions. Such decisions include

whether to invest in and/or make operational changes to reduce a generating unit’s

emission rate (if the emission rate is above the state’s performance standard); retire the

affected unit(s); increase output and overall plant utilization (if the emission rate is

below state performance standard); obtain and surrender emission allowances (at some

incremental opportunity cost) for each ton of CO2 emitted; or pay directly or indirectly

for investments in energy efficiency and/or zero-carbon energy resources to achieve

reductions in emissions.

EPA’s Clean Power Plan proposal has given the states broad discretion about how to

design their compliance plans. Ideally, states will elect to adopt plans that align with the

character of the electric industry structure affecting EGUs in different parts of their

states. This would mean that in regions with organized multi-state wholesale power

markets, states would proactively work with their counterparts in the region to develop

a common multi-state approach with similar core elements.

Why States Should Care About Aligning their CO2 Compliance Plans with

Wholesale Electric Market Structure

Recall that the motivation for restructuring the electric industry in many states and

regions was that electricity supply could be provided more efficiently through markets

than through traditional cost-of-service regulation.22 Changes in the industry structure

22 The long-standing goal of cost-of-service regulation of electricity supply has always been to mimic the

incentives produced in competitive markets through properly structured regulation. In theory, a regulatory

structure with such incentives to minimize cost and to incent innovation and good customer service will

provide for the efficient development and operation of generating facilities, meeting consumers’ electricity

needs at the lowest possible long-run costs. With the advent of potential competition in markets for power

generation in recent decades, many states moved to restructure their electric industries. At the same time,



coincided with significant changes in federal and state regulation of power plant

emissions, which began to rely on market-based mechanisms as the way to meet

environmental goals at lowest cost.23

As noted previously, the national experiment with the SO2 cap-and-trade program has

been widely viewed as a success for industry, states and EPA, and for consumers. This

market-based framework has provided emissions reductions that took advantage of the

innovative and cost-minimizing attributes of markets at significantly lower cost than

had been expected.

The design of the SO2-emissions-trading program has since been replicated in numerous

federal, regional and state programs in the U.S. and abroad. Notably, there are

numerous examples of multi-state cooperation on market-based environmental

programs, where such has been allowed under law.

A prime example of a multi-state market-based program is the “NOx SIP Call” program

to reduce emissions of NOx. That program had its roots in the efforts of the member

states of the Ozone Transport Commission (OTC)24 to adopt a region-wide budget (or

states and regions where industry restructuring took hold moved to create independent entities (today’s

RTOs) to ensure fair access to the transmission grid and administer open, competitive markets for electricity

supply. As previously discussed, the end result of these efforts is the competitive wholesale market

structures and entities that exist throughout much of the country. The benefits of markets flow from

moving investment and operational risks to those entities most capable of managing them and from the

pressure to reduce costs and improve quality in order to succeed and profit from market activities. In a

competitive market environment, pressure on market participants drives the industry to the lowest-cost,

most-efficient responses to changing customer needs. In the electric industry, this drives down wholesale

supply prices for electric ratepayers.

23 Similar economic incentives, principles, and responses are at work in the design and implementation of

market-based allowance-trading programs for control of power-plant emissions. As noted previously, some

environmental laws and regulations do not allow for market-based approaches and rely on unit-specific

technology-based pollution-control requirements, with significant administrative processes to administer

and enforce these standards. This approach is in some ways similar to the challenges of traditional public

utility regulation with prudency reviews: Air quality regulators attempt to determine administratively

what will be the most effective, “best-available-control-technology,” lowest-cost way to achieve air-quality

objectives, and establish requirements for individual power-plant owners based on this evaluation. But as

in the case of utility regulation, there are often practical constraints (e.g., limited or imperfect information,

work force and budget restrictions) on the ability of air regulators to do their jobs. Given these challenges,

the traditional unit-/technology-based air-quality framework is often viewed as a poor substitute for more

market-based and direct emission cap or pricing regimes, with emission-reduction solutions driven by the

decisions of market participants and priced by competitive market outcomes.

24 The OTC was established in the CAA Amendments of 1990 as a means to help states in the Northeast

work together to reduce emissions that affected the entire region’s air quality. Members of the OTC include

the District of Columbia and 9 states (Delaware, Maryland, Pennsylvania, New Jersey, New York,



limit) on NOx emissions. To control NOx emissions that move across state boundaries,

the OTC states developed a multi-jurisdictional cap-and-trade program. The resulting

OTC NOx Budget Program was the first cap-and-trade program formed by a group of

states and represents the first large-scale application of the cap-and-trade model to a

problem other than acid rain. Eventually, EPA expanded the number of states that

needed to address inter-state transport of NOx emissions through State Implementation

Plans (through the so-called “NOx SIP Call”), and provided a model rule that states

could elect to adopt to carry out an inter-state program.25 Eventually, all of the states

opted into the multi-state emission-trading program.26

Notably, the NOx SIP Call has parallels to today’s situation with the proposed Clean

Power Plan. The multi-state nature of the ozone and NOx-emissions issues and of the

industrial- and power-generation activities that contributed to the region’s air quality

problem lent itself to a regional solution. EPA acted to control a source of air emissions

that had impacts beyond state borders and had to use the State Implementation Plan

mechanism to do so. Individual states had the ability to develop a plan with command-

and-control elements to accomplish their emissions-reduction targets. But every state

eventually exercised its jurisdictional discretion to adopt a multi-state plan and to rely

on a market-based approach to create incentives for minimizing the overall cost of

compliance.

Connecticut, Rhode Island, Massachusetts, and New Hampshire) whose unhealthy air quality was affected

by the transport of emissions across state boundaries. http://www.epa.gov/captrade/documents/nox.pdf.

25 “The NOx Budget Program set a regional ’budget’ (or cap) on NOx emissions from electric power

generating facilities and industrial boilers from a variety of industry types during the ‘ozone season’ (from

May 1st through September 30th) beginning in 1999.… To meet the budget, sources were required to reduce

emissions significantly below 1990 baseline levels, and could use emissions trading to achieve the most cost-

effective reductions possible. At the end of each ozone season, sources must demonstrate that their actual

ozone season emissions do not exceed the amount of allowances held for that period. Unused allowances

may be sold or banked for use in a subsequent ozone season. Regardless of the number of allowances a

source holds, it may not emit at levels that would violate other Clean Air Act or state requirements. As with

any cap and trade program, sources can devise their own strategies to comply with NOx emission

restrictions. The ability to trade allowances places a value on emission reductions and encourages sources to

develop the most cost-effective emission reduction strategies to achieve the overall required emission

reductions. This approach allows the OTC states to achieve greater reductions than could be captured under

a traditional regulatory approach for the same overall cost.”

http://www.epa.gov/captrade/documents/nox.pdf 26 Eventually, the program included not only the original OTC states, but also parts or all of Missouri,

Illinois, Michigan, Indiana, Ohio, Kentucky, West Virginia, Virginia, Tennessee, North Carolina, South

Carolina, Georgia, and Alabama.

http://www.epa.gov/captrade/documents/nox.pdf



Similar motivations led to the development of the RGGI program, in which an original

group of 10 states sat down together to develop their own common approach to

reducing CO2 emissions from power plants within their borders. The states reserved for

themselves the ability to join the program – or not – and to adopt a framework that

would allow for cost-effective emissions reductions. Although the states’ memorandum

of agreement allowed each state to choose the manner in which it would distribute

allowances to affected power generating companies, the states eventually chose to

establish a program where there was little variation in the states’ implementation of the

program. 27 Eventually, all RGGI states joined into a central auction process as a means

to distribute allowances into the market and to establish the value of allowances.

The RGGI program is yet another example of a market-based emissions program that

has accomplished emissions reductions cost-effectively and in a manner that allows for

seamless implementation in centralized wholesale power markets.28 The program is

competitively neutral and affects the market only to the degree that a fossil-fuel power

plant has to obtain allowances to cover its CO2 emissions (if dispatched) and to build the

value of those allowances into its offer prices in RTO markets.

Market-based emissions-control programs like these provide critically important lessons

for EPA and the states as they move towards implementation of Clean Power Plan

requirements. Most importantly, the past two decades have demonstrated the natural fit

of market-based emission control approaches with competitive wholesale power

markets and the ability of these structures to support fair and efficient electricity market

outcomes while minimizing the costs of achieving mandated emission control

requirements. Absent this approach, conflicting and disjointed emission-control

incentives from balkanized state implementation will likely increase the cost of the

27 RGGI was a first-of-its-kind CO2 multi-state program with no prior framework. The states were able over

the course of just several years to work together to: (1) design the program with a focus on market-based

mechanisms and efficient design; (2) create the administrative rules, structures and organizations needed as

a region (e.g., RGGI Inc., an allowance trading platform, organizational bylaws and governance, market

monitoring procedures, etc.) and within states; (3) address the concerns, comments and perspectives of

affected entities and stakeholders; (4) create a state model rule; (5) agree upon the allocation of a regional budget

amongst states without having state-specific targets (as will be provided in the Clean Power Plan) and despite

fundamental allocation decisions that had very real financial implications for states and states’ asset owners;

and (6) implement enabling state rules, regulations and laws with full administrative proceedings in each

individual state.

28 See, for example: Paul Hibbard, Susan F. Tierney, Andrea M. Okie, Pavel G. Darling, “The Economic

Impacts of the Regional Greenhouse Gas Initiative on Ten Northeast and Mid-Atlantic States: Review of the

Use of RGGI Auction Proceeds from the First Three-Year Compliance Period, November 15, 2011; and Paul

Hibbard, Andrea Okie and Susan Tierney, “EPA’s Clean Power Plan: States’ Tools for Reducing Costs and

Increasing Benefits to Consumers,” July 14, 2014.



Clean Power Plan and diminish equity and efficiency in existing wholesale power

markets.

Benefits of Market-Based Approaches in States’ Clean Power Plans

In its proposed Clean Power Plan, EPA has recognized many elements supporting the

states’ ability to adopt plans that rely on market mechanisms and to establish these plans

in ways that respect the regional nature of power production in most parts of the U.S.

Under Section 111(d), EPA must give states the ability to craft their own plans to comply

with their emission-reduction targets. But beyond that, EPA has given the states the

option to develop multi-state plans that align with the geographic realities of interstate

wholesale power markets. This is an opportunity that invites states to afford themselves

of the benefits of regional collaboration and market-based environmental strategies.

There are many benefits of adopting common market-based compliance plans that align

well with competitive markets and avoid the potential for market distortions that could

arise through inconsistent and uncoordinated state plans.

Ideally, such a system would have states individually electing to adopt a common

framework that includes a mass-based cap covering a multi-state region, providing for

emissions-averaging across the footprint of the system as a whole and folding in new

units into the system as soon as they enter commercial operation (or shortly thereafter).

The benefits of such an approach include:

- Economic efficiency of power supply: A multi-state regional power market with

an aligned multi-state CO2 emissions program would create transparent price

signals and appropriate economic incentives to owners of generating resources

and suppliers of demand-side services to provide electricity services as efficiently

as possible, taking into account carbon emissions and avoided carbon emissions.

Such a system would allow for the translation of a marginal cost of compliance –

the cost to emit the next ton of CO2 – into part of the variable cost of electricity to

be included in offer prices into wholesale electricity markets. This permits the

grid operator to continue to implement economic dispatch of electric resources

on a fair and efficient basis across all resources – regardless of fuel, age, or

efficiency. This expands the universe of compliance options to include not only

trading of emission-reduction opportunities across affected sources, but also the

“trading” of emission reduction opportunities across affected and non-emitting

sources (e.g., hydro, other renewables, nuclear, energy efficiency). Expanding

the options for compliance will reduce the total costs of compliance relative to



the alternative. A multi-state compliance program would also reduce the market

distortions and unintended outcomes that may result from a patchwork of

different state programs.

- Lower-cost compliance: As noted previously, experience with market-based

emissions-trading approaches indicates that overall environmental compliance

costs of emissions-trading programs are lower than original estimates and lower

than alternative command-and-control programs. Recent modeling of multi-

state market-based approaches indicates the economic advantages of such an

approach relative to single-state and/or non-market-based approaches from a

cost-of-compliance point of view. Such modeling has been conducted by the



Bipartisan Policy Center, for example,29 and by PJM with inputs from state

regulators on the set of scenarios to analyze.30

29 Modeling by a team from the Bipartisan Policy Center (BPC) reached a number of conclusions, including

these (quoted from the BPC presentation):

The magnitude of impacts from the Clean Power Plan (CPP), including potential compliance costs, are

dependent on EPA and state decisions yet to be made, as well as market factors, such as: the availability

of end-use energy efficiency (EE); the price of natural gas, and the future of existing nuclear plants.

This uncertainty increases the value of policy designs that inherently create the incentives for

implementing least-cost solutions and allow affected companies flexibility to adapt to changing

circumstances.

Interconnected nature of the power system is important to consider when looking at costs and impacts

of Clean Power Plan

Benefits of multi-state collaboration and/or linked trading approaches: Adopting policy designs that

allow access to emission reduction opportunities in other states tends to significantly lower the cost of

compliance and reduce retirements.

State choice of energy efficiency policies will significantly impact the cost: Effective end-use energy

efficiency policies are important for cost containment. Demand reductions dramatically reduce system

cost because they both reduce the need for additional capacity and lower fuel costs due to reduced

demand.

Treatment of new builds is an important policy consideration: Including new sources in

implementation policies reduces potential market distortions and tends to lower cost; Different

implications depending on state choice of rate-or mass-based goals.

State policy choices will impact generation mix, investments, cost, and CO2 emissions. Choice of rate-

or mass-based goals and implementation policies: Mass-based implementation tends to lower total

cost, while rate-based implementation has less impact on wholesale electricity prices. Despite projected

wholesale electricity price increases in some states/scenarios, end-use EE may keep customer bills from

increasing. Mass-based policies limit generation shifts and emissions leakage between states.

Jennifer Macedonia, Blair Beasley, Tracy Terry, Meghan McGuinness, and Stuart Iler, “Insights from

Modeling the Proposed Clean Power Plan,” Bipartisan Policy Center, April 2015.

30 PJM recently conducted analyses of the changes in system-wide production costs assuming various

designs of states’ compliance plans. Quoting from the PJM report, the “high-level insights from the

economic analysis include:

Fossil steam unit retirements (coal, oil and gas) probably will occur gradually. As the CO2 emission

limits decline over time, the financial positions of high-emitting resources should become increasingly

less favorable, with lower-emitting resources displacing them more often in the competitive energy

market.

Electricity production costs are likely to increase with compliance because larger amounts of higher-

cost, cleaner generation will be used to meet emissions targets.

The price of natural gas likely will be a primary driver of the cost of reducing CO2 emissions if natural

gas combined-cycle units become a significant source of replacement generation for coal and other fossil

steam units.

Adding more energy efficiency and renewable energy and retaining more nuclear generation would

likely lead to lower CO2 prices; this could result in fewer megawatts of fossil steam resources at risk of

retirement because lower CO2 prices may reduce the financial stress on fossil steam resources under this

scenario.



- Reliability: A common multi-state market-based program will provide the

system operator with maximum flexibility to continue to operate the system with

security-constrained economic dispatch, taking into consideration the carbon

intensity of power sources as well as the essential reliability services needed to

maintain the integrity of the grid. The character of the multi-state market will act

as a form of Reliability Assurance Mechanism and complement the reliability

tools available to the grid operators.31

- Administrative simplicity and efficiency for assuring compliance: While the

initial set up of emission- and allowance-tracking systems requires an up-front

expense, a cap-and-trade program is easier to administer than a program

implemented through unit-by-unit control technology and unit-specific permit

restrictions. Economic incentives built into the program to ensure that an EGU’s

emissions are covered by the surrender of a sufficient number of allowances (and

with financial penalties for failure to do so) support ease of administration. The

transparency of the emissions-trading system and of the performance of the

system as a whole avoids the administrative complexity of measurement and

verification systems associated with energy efficiency programs. From the point

of view of the energy-efficiency measures’ impacts on carbon emissions of the

system, those impacts end up being embedded into the outcomes of the dispatch

of the electric system. The same is true for incorporating the value of zero-

State-by-state compliance options, compared to regional compliance options, likely would result in

higher compliance costs for most PJM states. This is because there are fewer low-cost options available

within state boundaries than across the entire region. However, results will vary by state given

differing state targets and generation mixes. PJM modeled regional versus individual state compliance

only under a mass-based approach.

State-by-state compliance options would increase the amount of capacity at risk for retirement because

some states likely would face higher CO2 prices in an individual compliance approach.”

“PJM Interconnection Economic Analysis of the EPA Clean Power Plan Proposal: Executive Summary and

Frequently Asked Questions” March 2, 2015, included as an attachment to the statement of Michael J.

Kormos, Executive Vice President – Operations, PJM Interconnection, before the Federal Energy Regulatory

Commission, Docket No. AD15-4-000, “Technical Conference on Environmental Regulations and Electric

Reliability, Wholesale Electricity Markets, and Energy Infrastructure,” March 11, 2015.

31 We discuss such tools in our papers: Susan Tierney, Paul Hibbard and Craig Aubuchon, “Electric System

Reliability and EPA’s Clean Power Plan: Tools and Practices,” February 2015; Susan Tierney, Paul Hibbard

and Craig Aubuchon, “Electric System Reliability and EPA’s Clean Power Plan: The Case of PJM,” March

16, 2015; and Susan Tierney, Eric Svenson, Brian Parsons, letter and report to Chairman Norman Bay, FERC,

re: Ensuring Electric Grid Reliability Under the Clean Power Plan: Addressing Key Themes from the FERC

Technical Conferences, Docket No AD15-4, April 17, 2015.



carbon generation, as well. “A mass-based approach captures all emissions

reductions that occur at the covered plants, whatever the reason for those

reductions, without the need to design and implement a crediting mechanism for

those reductions. Importantly, reductions can be captured from activities or

events that EPA or a state might not allow a state to credit in the rate-based

context, or that may be difficult to credit.”32

- Equity: Such a system would treat all emitting units – new or existing, efficient or

inefficient, coal or gas – fairly from the point of view of their CO2-emissions

compliance burden by subjecting them on the margin to the same emissions-

reduction cost per ton of emissions, thus enabling the accomplishment of the

overall emissions-reduction goal at the lowest overall cost.

- Environmental integrity: Once emissions caps are established, the system’s

performance can be met, verified, and enforced. Continuous emissions

monitoring leads to highly accurate emissions monitoring. Emitters are

generally motivated to comply by punitive sanctions if they fail and/or by

requirements to make up for excess emissions at some multiple above

exceedances.

In sum, and most importantly from the perspectives of market impacts and consumer

costs, emissions-cap and allowance-trading programs drive lowest-cost emission

reduction outcomes. They can be seamlessly integrated into competitive wholesale

power market mechanics in a fair and efficient way. Carrying out such programs

through cooperation among states in a regional RTO market will support those regions'

continued co-optimization of cost and reliability outcomes.

32 Franz Litz and Jennifer Macedonia, “Choosing a Policy Pathway for State 111(d) Plans to Meet State

Objectives,” April 14, 2015. http://bipartisanpolicy.org/wp-content/uploads/2015/05/Policy-Pathways-

Paper.pdf. The authors say further that in a rate-based approach, “In order to credit emissions reductions or

avoided emissions that result from activities outside the fence line of power plants—such as through energy

efficiency or renewable energy projects—a state must design and implement a crediting mechanism for each

type of credit. This is the biggest administrative challenge in the ratebased context that does not exist in the

mass-based. Some eventualities that reduce emissions may not affect the emissions rate, such as plant

retirements or when demand is reduced for reasons that cannot be credited. In addition, some have raised

concerns that credits and the crediting process can be legally challenged, including through citizen suit

actions.”

http://bipartisanpolicy.org/wp-content/uploads/2015/05/Policy-Pathways-Paper.pdf

http://bipartisanpolicy.org/wp-content/uploads/2015/05/Policy-Pathways-Paper.pdf



4. CONCLUSION

We strongly encourage states with electric generating units in organized wholesale

power markets to coordinate the development of state plans so that they can elect to

participate in a common, mass-based multi-state emissions-trading program that covers

both existing EGUs and new NGCCs as they enter the

market over time.

We hope that states will decide for themselves that this is

the pathway toward a least-cost compliance strategy for CO2

emissions reductions and an efficient, reliable power system

for the benefit of their consumers.

In saying that, we do not mean to suggest that states would

decide to eliminate other energy programs affecting the

power sector, including renewable portfolio standards,

energy efficiency programs, clean energy standards, or

other policies. We think that a well-designed multi-state

mass-based system dovetails with these other policies, and we note and respect the

findings of so many analysts (including our own work) that indicate that overall cost of

compliance will be lower with energy efficiency as a lead strategy.

Although we do not expect that EPA’s final rule will impose a requirement that states

adopt the type of multi-state, mass-based plan we urge them to support, we nonetheless

encourage EPA to provide even greater incentives for states to adopt mass-based

approaches that consolidate new NGCCs into their program for existing generating

units.

Consistent with comments filed by many parties in EPA’s rulemaking docket for 111(d)

and/or 111(b), there are many examples of provisions that EPA could incorporate in the

final rules (and/or preamble) to encourage states to move in this direction, including:

- Provision of a default/presumptive mass number for each state expressed

alongside the rate target.

- Streamlined review of State Plan submittals that include a multi-state mass-based

program covering generating units that operate in an RTO, including release

from a requirement to submit a plan revision with each new NGCC that enters

operation in the state or release from a requirement to adopt a memorandum of

We encourage states with

EGUs in RTO markets to

adopt a mass-based

emission-trading

program that covers both

existing and new NGCCs

in the footprint of the

RTO. And we encourage

EPA’s final rules to

include incentives for

states that adopt such

approaches.



understanding with other states with a mass-based approach in order to trade

with them.33

- Preparation of a model rule or a national tracking system/platform that would

allow for the creation and trading of CO2 emission allowances and/or emissions-

reduction credits.34

- Providing states more time to submit their States Plans if they can show that they

are working toward the development of a multi-state approach.

- Introduce detailed administrative requirements, accounting rules, and ‘burdens

of proof’ in instances where a state builds a plan based on a rate-based approach

and wants to trade emissions-reduction credits from a state with a mass-based

approach, and vice versa.35

33 Comments of Calpine Corporation: “The New Source/Existing Source Coverage Gap is Potentially the

Most Significant Shortcoming of the Proposed Clean Power Plan…..Failure to address this gap in coverage

could result in serious market distortions and higher costs for ratepayers than are necessary to achieve the

Proposed Rule’s goals. …EPA’s publication of presumptively approvable mass emissions budgets that

include new sources is a positive step in the direction of incenting states to cover new sources in their state

plans. EPA should do all it can to encourage states to cover new sources in their state plans.” Calpine

Corporation Comments on the Proposed “Carbon Pollution Emission Guidelines for Existing Stationary

Sources: Electric Utility Generating Units,” Docket ID No. EPA–HQ–OAR–2013–0602, page 4.

34 Comments of the Clean Energy Group: “We urge EPA to support the development of credit markets for

111(d) compliance. Such markets should function similar to existing [renewable energy credit] REC markets,

which verify and track credits to avoid double counting, and states may elect to utilize the existing REC

market as a component of their compliance plans to the extent that RECs have equivalent attributes. We

recommend that EPA issue guidance for state plans that clarifies the use of compliance markets as a part of

state plans. For example, we expect that some states may be interested in submitting partial joint plans that

reference the use of a national 111(d) energy credit for compliance without entering into memoranda of

understandings with each state that might be a generator or purchaser of renewable energy credits, and

EPA should ensure it does not create a barrier for this compliance option. If requested by states, EPA may

also want to consider developing a national tracking system for 111(d) compliance credits to facilitate this

multi-state dynamic.” Clean Energy Group Comments on the Proposed “Carbon Pollution Emission

Guidelines for Existing Stationary Sources: Electric Utility Generating Units,” Docket ID No. EPA–HQ–

OAR–2013–0602 (hereafter “Clean Energy Group comments”), page 11.

35 Exelon Corporation’s comments suggest various accounting rules to accompany trades among EGUs in

states with dissimilar emissions-reductions schemes. Such “rules are designed to count each emission

reduction once and only once and can also be summarized as a set of eight accounting rules:

1. When a mass-based state sells a CO2 allowance, the selling state’s generators must reduce their

emissions by one ton.

2. When a mass-based state buys a CO2 allowance, the buying state’s generators may emit one more ton

of carbon.



- Clarify the value of a mass-based approach that coincides with the borders of

RTO regions as an explicit tool for assuring electric system reliability.

- Provide a disincentive for a rate-based state backsliding on existing zero-carbon

resources that are not factored into the rate target, in light of the consequences

for overall CO2 emissions with loss of such resources.

- Address the concern of some observers that a mass-based approach constrains

growth, perhaps through allowing a protocol to adjust the mass-based cap with

the addition of new generating units satisfying NSPS, or explaining how the

addition of energy efficiency or zero-carbon resources adds no emissions but

reduces price pressure on emissions allowances.

In sum, we encourage EPA and the states to:

Seek consistency in the design of Clean Power Plan compliance approaches across

states within integrated wholesale market regions. The broader the geographic scope

of compliance, the more efficient, lower cost, and more reliable the outcome.

Seek consistency in the manner in which the burden of CO2-emissions reduction falls

on different classes of units: as much as possible, encourage the creation of a system

that is blind to age and type of generating technology, and instead assigns a

consistent compliance cost per unit of energy produced so that the social costs of

compliance are minimized.

With such principles, there is much greater likelihood of outcomes beneficial to

consumers: efficient electricity markets and efficient environmental-compliance

approaches.

3. When a rate-based state buys a CO2 allowance, the buying state may deduct one ton of carbon from the

numerator of its compliance calculation.

4. When a rate-based state sells a REC, the selling state may not take credit for the REC or the underlying

zero-carbon generation.

5. When a mass-based state that does not use allowances sells a REC, the selling state must reduce its

emissions budget by 0.5 tons (assuming that the mass value of a REC is 0.5 tons of CO2).

6. A mass-based state that uses allowances may not sell a REC to be used for Clean Power Plan

compliance in another state.”

Exelon Corporation, Comments on the Proposed “Carbon Pollution Emission Guidelines for Existing

Stationary Sources: Electric Utility Generating Units,” Docket ID No. EPA–HQ–OAR–2013–0602, page 103.



APPENDIX: PRIMER ON ORGANIZED WHOLESALE POWER MARKETS

The specific design of electricity markets across the RTOs varies somewhat, but in all of

them, the pricing for electric energy (in MWh) is established based on the offer price of

the marginal generating unit dispatched to meet load at a particular time (e.g., an hour).

This is called the clearing price. Owners of power plants tend to offer their electric

energy supply at a price that roughly reflects their costs of operation. Essentially, the

clearing price is paid to all generators that are dispatched to provide physical supply of

power, and generators that are not ‘cleared’ (selected) receive no compensation in the

energy market in that time period.36

In terms of longer-term signals to spur investment when needed, certain RTOs (ISO-NE,

NYISO, and PJM) administer fully competitive auctions for procuring capacity. Other

RTOs (CAISO, MISO) have a hybrid system combining market procurement of capacity

with self-supply obtained through ownership or bilateral contracts. ERCOT and SPP

have no capacity markets. In all regions, the RTOs procure a variety of things (called

ancillary services37) needed for reliable system operations through some mix of market-

based and non-market mechanisms.

36 In practice, RTOs tend to have “day-ahead” and “real-time” markets for energy. Most binding financial

commitments are made in the day-ahead market, whereby generating asset owners offer to supply energy at

specified prices in each hour of the subsequent day. If selected to operate – or “committed” – a resource is

paid in each hour of the next operating day based on the offer of the last (marginal) unit selected to operate

in that hour to cover energy demand. Real-time markets settle out differences between load and generation

close to real time throughout the operating day (e.g., if load is higher or lower than expected in the day-

ahead market).

37 Ancillary services include a variety of products and services (e.g., spinning reserves, automatic generation

control, regulation service, black-start service). Typically they can be supplied by certain dispatchable loads

and/or power plants with certain operational characteristics (e.g., fast-start and ability to ramp up or down

quickly). The RTOs’ hourly markets also generally support the provision of ancillary services that are a

function of hourly load and generation, such as reserves (having some quantity of generating capability

above actual electrical demand in the market and available to operate in order to be able to “fill in” in the

event of a sudden and unexpected loss of generation or other system assets); and “regulation” or “load-

following” service (having resources on the system with room to quickly increase and decrease output in

order to manage second-to-second and minute-to-minute variations in net load on the system). Since the

requirements for these services is also hourly and tied to generation capability, payments and pricing for

such services is generally managed in combination with energy market unit commitment and dispatch

methods.



In the wholesale market regions, owners of non-utility-owned generating assets must

recover their costs of investment and operation through markets,38 rather than through

regulated rates. Over the operating life of the asset, a resource owner seeks to recover

total revenues – across all central market and bilateral market activity – in amounts that

cover their operating costs, their investment, and an acceptable return on that

investment.

Operations and Incentives in Energy and Ancillary Service Markets

At a minimum, energy/ancillary service markets allow power plant owners to recover

their costs of producing energy in those hours. This is because the minimum offer price

by a generation owner will include at least all of the variable costs of operation,

including, for example, fuel costs, variable operations and maintenance (O&M) costs,

and emission costs (e.g., the variable cost to operate emission control systems, or the cost

to purchase allowances to emit the quantity of pollutant emitted in that hour). This

amount will be all the asset owner recovers only if it is the marginal unit dispatched in

that hour (meaning the clearing price paid to all units is exactly equal to this marginal

unit’s offer), and the offer price exactly equals these variable expenses. In this

circumstance, then, the unit would be able to cover its operating expenses, but it would

earn no additional revenue to offset fixed/annual costs or provide a return of or on

investment.

In reality, most generating units that end up being dispatched typically earn some

margin above their variable operating expenses. This may be due in part to the ability of

an asset owner to build into its offer some amount of cushion above expenses.

Moreover, over the course of a year the unit will earn revenues from the energy market

in excess of variable costs because in many hours the unit will not be the marginal unit,

but rather will be “inframarginal,” and that the clearing price it earns (set by a higher-

priced unit on the margin) will exceed its costs. In some hours this difference can be

large.

Although RTOs seek to dispatch plants in economic-merit order (starting with the

lowest cost and then committing plants up to the amount needed to serve load and

provide ancillary services) and in so doing minimize the total costs to provide power

38 “Markets” reflects a combination of (a) centralized markets for the physical supply of energy, ancillary

services, and (in some RTOs) capacity, and (b) bilateral financial transactions (e.g., hedging contracts; joint

ownership agreements) in which the seller and buyer make commitments that lock in price certainty and

other obligations surrounding electricity supply.



over time, there are features of the electric system that constrain this goal in practice.

These constraints include, among other things, the operating characteristics of power

plants or the existence of congestion on transmission pathways that prevent an

economical unit from being operated.

RTOs dispatch the system fully aware of unit-specific information about the operating

characteristics of the power plants on the system, regardless of who owns them. Each

generating unit provides information about such things as: the time it takes to start up

the unit and then ramp it up to full load; operational limits between maintenance

outages; the minimum load at which it can operate; the number of starts and stops it

may have in a year; and so forth. Some generating units have limitations in their air-

emissions permits that set out their maximum run-time on certain fuels. The RTO takes

all of these specifications into account in determining the dispatch of the system. These

restrictions also limit a generating unit’s ability to operate and capture revenues in

wholesale markets, and can lead to the dispatch of a higher-priced unit on the margin.

Incentives in Capacity Markets and Resource Adequacy Obligations

Typically, if a developer of a new power plant or the owner of an existing one expects

that net energy and ancillary service market revenues will equal or exceed what the

developer/owner needs to cover going-forward capital investment plus a return on that

investment, that developer/owner will proceed with the investment absent any

additional sources of revenue. However, in most RTO regions, energy market prices are

not typically high enough or often enough to produce sufficient net revenues to support

major capital investment – whether in new or repowered generating capacity, or in

major incremental investments in existing assets.

Consequently, many RTOs either administer central markets for the purchase of

capacity or set a “resource adequacy” requirement on load-serving entities which

provide incentives for them to construct or enter into bilateral arrangements to purchase

capacity from generators (and/or demand-side resources). In some cases the RTO

administers some form of hybrid approach that has elements of both a capacity market

and a resource adequacy/self-supply requirement. These frameworks are designed (in

combination with net energy/ancillary service market revenues) to provide incentives

for the construction of new capacity resources when needed (as a result of load growth

or retirement of existing assets), and/or to provide sufficient revenues for investment in

unit upgrades, repairs, or other capital-intensive requirements (such as add-on emission

control systems). Specifically, these mechanisms provide for an additional potential

stream of revenues to generators.



Importantly, the mechanism and timing of capacity market revenues is significantly

different from energy, with implications for how resource owners think about how they

will manage environmental compliance options and whether they will make

investments and/or operational changes to deal with environmental obligations. Energy

prices are set and change on an hourly basis, and are priced and collected on the basis of

dollars per megawatt-hour of energy generated. Capacity payments are typically fixed

monthly payments based on prices determined in monthly or annual auctions.39

The incentive for investment in new capacity and for investment in major capital

additions on an existing unit – e.g., for the installation of emission control technology –

is driven in part by a developer/owner’s expectations of net revenues to be earned in the

various wholesale markets. If expected net revenues equal or exceed what is needed to

make the investment with a sufficient return on that investment, no additional revenue

streams – such as from a capacity market – are needed to decide to go forward with the

investment.

39 For example, in New England and PJM, capacity auctions occur three years prior to the year of need, and

result in a price per unit of capacity (e.g., dollars per kilowatt-month) that is paid out to suppliers on a

monthly basis.

Analysis Group - Carbon Control and Competitive Wholesale … · 2018-10-05 · Carbon Control and...

Documents

Transcript of Analysis Group - Carbon Control and Competitive Wholesale … · 2018-10-05 · Carbon Control and...